The purpose of the proposed research is to study the effectiveness of the drug court model in comparison with treating drug-using offenders through more traditional probation-based services. The research will compare drug court with traditional court processing of drug involved offenders in five Florida judicial circuits and the subsequent management of these cases by probation and treatment services. The research will characterize the linkages among and the relative effectiveness of nine judicial processing/supervision/treatment modalities, measure individual-level participation and outcomes, and assess the cost-effectiveness of the alternative models. The proposed research builds on previous research we have conducted on drug treatment effectiveness. Major study components are: Observation of drug courts and probation to identify the linkages between and the ways in which these linkages differ in circuits with operational drug courts. Secondary data analysis that will expand on previous research to examine specific differences in recidivism outcomes between those monitored by drug courts and those not monitored by drug courts conditioned on the type of supervision and the type of treatment; these analyses will consider within-circuit and between-circuit differences. Interviews with drug court and comparison group individuals at baseline (time of admission to probation), and at 3-months, 6-months and 12-months post-baseline to collect additional information. Hair assays (tests) conducted at 12 months' followup to identify drug use among individuals assigned to the different supervision/treatment conditions. Administrative record data collection and analysis at 12 post-treatment months to identify criminal recidivism outcomes (arrests, new sentences to probation or prison, revocations). Cost-effectiveness analyses to compare the marginal value added of drug court supervision to two different levels of probation supervision. The study will execute several process/qualitative components during the first 2 years, and the outcome/quantitative evaluation during Years 3 through 5. We expect that the study will yield approximately ten peer-reviewed journal articles beginning with two process evaluation articles in the second year of the study.